Analysis of uncertainty and variation in underground train-induced vibration based on measured data

Ground-borne noise and vibration from metro trains can cause human distress/annoyance, affect the growth of animals and plants, and also negatively affect real estate property values. Before construction, the common prediction methods for the environmental vibration caused by metro trains are physical model prediction and analogy test. However, the rationality of the analogy test and the factors that need to be considered in the simulation prediction have not been systematically tested and studied. In order to solve this problem and provide experimental reference for the prediction model, one of China's subway were systematically tested to investigate wheel roughness, rail roughness, train running speed, axle load changes, the vertical and lateral contributions, and the influence of random factors existing in the test on the vibration of subway operating environment, the following conclusions are obtained: (1) The test state randomness (random factors in the test, such as wheel-rail contact points, roughness phase, etc.) mainly affect frequencies below 31.5 Hz segment, mainly from the randomness of the wheel roughness phase or the variation of the spatial position of different contact points during actual operation, which result in large different of long-wave roughness of the wheel and rail; (2) The wheel-rail lateral force also has great influence on the environmental vibration. When establishing the environmental vibration prediction model, it is necessary to consider the state that the train reaches the straight line after passing through the curve, and the influence of track lateral irregularity; (3) The impact of wheel roughness on the environmental vibration is very large. Basically, it has an impact on the whole evaluation frequency band of environmental vibration.